Prostate cancer is extremely prevalent, striking 1 in 6 American men. Current methods to identify it are limited. Biopsy is not performed in all men at risk and misses cancer in 30-50% of its cases. Once diagnosed, a man faces the dilemma of choosing aggressive therapy, with its undesired side effects, or risking metastases and death. While Gleason Grading of biopsy samples can provide an indication of metastatic risk, biopsy frequently misses the more aggressive components of a cancer. Clearly there is an important need to improve prostate cancer identification and characterization of aggressiveness. Magnetic resonance imaging has shown promising results in detecting and characterizing prostate cancer; however it is still limited. Our recent studies have suggested a tissue specific contrast mechanism in DCE MRI and diffusion MRI that we plan to exploit to improve cancer identification and characterization. In particular, our results suggest that the MR contrast agent (Gd-DTPA) enters cancerous glands and ducts but not healthy ones. In preliminary data, we have shown DCE MRI differences between aggressive prostate cancers (Gleason Score 4+4) and moderately aggressive prostate cancers (Gleason Score 3+3) that are consistent with this hypothesis. Current analyses of DCE MRI data do not take this tissue behavior into account. The goal of the current work is to design a nonlinear, mixed effects statistical model that can improve identification and characterization of prostate cancers based upon combining DCE MRI analyzed with pharmacokinetic models that incorporate Gd-DTPA impenetrable glandular lumen and diffusion MRI. We hypothesize that such a model will improve the identification of prostate cancer and provide discrimination among Gleason Grade 3, 4, and 5 prostate cancers. This proposal aims to: I. Determine if a novel pharmacokinetic model incorporating luminal water better discriminates prostate cancer tissues than standard modeling, II. Determine if a novel, nonlinear mixed effects statistical model combining DCE MRI and Diffusion MRI has higher sensitivity in identifying prostate cancer in both peripheral zone and central gland than standard models, and III. Determine if a novel, nonlinear mixed effects statistical model of DCE MRI and Diffusion MRI can discriminate among different Gleason Score cancers and non cancerous tissues. We propose to exploit a novel contrast mechanism identified by our research group based on a combination of DCE and diffusion MRI findings which could potentially improve detection of cancer and more importantly discriminate high grade disease that needs to be treated aggressively from low grade, indolent disease, which would allow the patient to delay aggressive therapy and its side-effects.